毕业论文移动版

摘要近红外光谱技术现在在石油、农业、制药等领域质量控制中得到普遍发展， 不仅能够进行定量分析，还可以实现在线生产监控。近红外光谱技术具有快速、 高效、无损、消耗溶剂少等优点。目前检测毒死蜱含量的方法主要是高效液相 色谱和气相色谱等方法。本文对近红外光谱技术检测溶液中毒死蜱含量进行了 研究。运用近红外光谱技术结合偏最小二乘法对溶液中微量的毒死蜱含量进行测定。用标准物质质量浓度为 1mg/kg 的毒死蜱及蒸馏水配制 39 个质量比范围 为 0。005~0。1mg/kg 的样本，其中蒸馏水作为稀释液对标物进行稀释。所有样本 按照浓度升序排序，依 2:1 分为校正集和预测集，其中浓度最大和最小样品应为校正集，预测集样品浓度范围应被校正集样品范围所包含，其中 26 个样品作 为校正集，13 个样本作为预测集。试验结果表明，选取 1100~1500nm 波长范围 的光谱，用二阶导数（2nd-der）结合标准正态变量变换（SNV）方法进行预处 理，采用主因子数 8 进行建模可得到最佳结果。此时，在校正集（留一交叉验 证法）取得 99。61%的预测准确率，预测集有 99。39%的预测准确率，校正标准差 为 0。00176mg/kg，预测标准差为 0。00240mg/kg。模型中预测值与化学值之间具 有显著的线性相关性。由此可知，近红外光谱技术可用于毒死蜱残留的测定。80724

毕业论文关键词：毒死蜱；近红外光谱技术；Matlab；PLS

Abstract Near infrared spectrum technology has been widely used in petroleum, in petroleum,agriculture,and other fields of quality control,can be either quantitative analysis,also can undertake online monitoring of production。Near infared spectrum technology is fast,efficient,non-destructive and low solvent consumption,etc。The level of detection of chlorpyrifos,at present is mainly high performance liquid chromatography(HPLC)and gas chromatography and other methods。In this paper,the near-infrared spectroscopy to detect pesticide chlorpyrifos content in solusion was studied。Using near infrared spectroscopy combined with partial least squares method for solution of trace organophosphorus pesticide chlorpyrifos were determined。With standard substance mass concentration of 1mg/kg of chlorpyrifos and distilled water preparation 39 quality than the range of 0。005~0。1mg/kg of samples,including distilled water as the dilluent of the standard dilution。All samples according to the concentration ascending order,in accordance with the2-1 is pided into calibration set and prediction set,and the concentration of maximum anf minimum sample for calibration set,sample concentration range prediction set range should be calibration samples contained,of which 26 samples as a calibration set,13 samples as  a prediction set。The test results show that the selection of 1100~1500 nm wavelength range of the spectrum,with the second derivative combined with standard normal variable transformation (SNV) method for pretreatment,using  principal factor number 8 model can get the best results。At this point,the calibration set (leave a cross-validation method) to obtain 99。61% of the prediction accuracy and predictionset 99。39% of forecast accuracy,calibration standard deviation accuracy is 0。00176mg/kg,prediction standard deviation is 0。00240mg/kg。Model has signficant linear correlation betweenpredicted values and the chemical   value。Accordingly,near

infrared spectrum technology can be used in the determination of chlorpyrifos residues。

Keywords: chlorpyrifos;near infrared spectroscopy technology;Matlab; PLS

目 录

第一章 绪论 1

1。1 毒死蜱 1

1。1。1  毒死蜱危害 (责任编辑：qin)